A point-to-multipoint (“P2MP”) link is a specific type of multipoint link which consists of a central connection endpoint (“CE”) that is in direct communication with multiple peripheral CEs. A point-to-multipoint link may be viewed as a tree structure having a root node in direct communication with a plurality of leaf nodes, wherein the central connection endpoint, or backbone edge bridge (“BEB”) is represented as the root node and the peripheral CEs (also BEBs) are leaf nodes. Any transmission of data that originates from the central CE is received by all of the peripheral CEs while any transmission of data that originates from any of the peripheral CEs is only received by the central CE. The leaf nodes cannot communicate with each other directly, but may communicate through the root node.
Examples of communication networks using point-to-multipoint links may include a corporation having a “central” headquarters and a number of off-site locations with each off-site location communicating with the headquarters over a corporate intranet. Other examples may include a financial institution, such as a bank, having a home office and several “branch” offices with each branch communicating back to the main office through a dedicated path. In each instance, one facility, location or server acts as a root node, serving as a “go-between” for the entire network.
Often, communication networks include a “working” path and a “protection” path. In some networks, the path has a tree structure. The protection tree includes a series of redundant core nodes and links which provide an alternate route between the root node and the leaf nodes. Ideally, the protection tree is route diverse, i.e., both node diverse and link diverse. Thus, if a link in the working tree should fail, traffic may be routed to the protected tree and vice versa.
Currently, IEEE standard 802.1Q-2005 amendment 802.1Qay is before the Institute of Electronic and Electrical Engineers (“IEEE”) standards committee. The new amendment defines a solution for point-to-point trunk protection switching only, even though the amendment also defines a point-to-multipoint trunk structure. Currently, there is no in-band signaling solution for coordinating the protection switching of the endpoints of a point-to-multipoint open systems interconnection (“OSI”) layer 2 tree topology.
Additionally, certain provisions of the point-to-point protection switching protocol proposed in amendment 802.1Qay would be impractical or impossible to implement on a point-to-multipoint basis. For example, management system requests originating from an operator are currently routed to both ends of a point-to-point trunk under the proposed standard. In the context of a point-to-multipoint system containing n leaf nodes, this type of implementation would require n+1 requests to be processed through the system. For large values of n, this approach needlessly bogs down the system and requires considerably more time to allow for switching, thereby increasing the likelihood that some leaf nodes may not be on the same link, i.e. working or protection, at any given time. Such a scenario is undesirable for network operators as it results in a greater chance of maintenance actions inadvertently causing undelivered messages.
Therefore, what is needed is a system and method for in-band signaling for use in point-to-multipoint trunk protection switching at the packet layer.